Add "quantization" as option to Vector stores

XMLWordPrintableJSON

    • Type: Task
    • Resolution: Unresolved
    • Priority: Unknown
    • None
    • Affects Version/s: None
    • Component/s: None
    • Python Drivers
    • Hide

      1. What would you like to communicate to the user about this feature?
      2. Would you like the user to see examples of the syntax and/or executable code and its output?
      3. Which versions of the driver/connector does this apply to?

      Show
      1. What would you like to communicate to the user about this feature? 2. Would you like the user to see examples of the syntax and/or executable code and its output? 3. Which versions of the driver/connector does this apply to?
    • None
    • None
    • None
    • None
    • None
    • None

      Context

      InĀ How Vector Search Scales by henry.weller@mongodb.com, a new "quantization" field is shown added the to our "vector" indexes. See below. In order for this quantization to work, will the "path" have to be a binary vector subtype? If so, I imagine that the binary values will already have to match the type specified in the quantization, right? To my knowledge, we don't have this work on our roadmap, not even for our Langchain repo.

      Definition of done

      This task is for the quantization option, to expose it, and create indexes with it, like so:

      { "fields": [\{ "type": "vector", "path": "plot_embedding", "numDimensions": 1536, "similarity": "cosine", "quantization": "none" | "scalar" | "binary" }

      ] }

      Pitfalls

      These indexes might not be ready yet on Atlas (cloud and local.

            Assignee:
            Unassigned
            Reporter:
            Casey Clements
            Votes:
            0 Vote for this issue
            Watchers:
            1 Start watching this issue

              Created:
              Updated: